Research On Unmanned Vehicle Positioning And Mapping Method Based On Laser SLAM

In recent years,Simultaneous Localization and Mapping(SLAM)technology has been widely used.Especially in the field of unmanned driving technology,laser SLAM has become a necessary method for unmanned vehicles to perceive the environment.Based on the application of laser SLAM in the field of low-speed unmanned driving,the research on the construction and positioning method of the unmanned vehicle navigation map was carried out,and the positioning mapping of the complementary fusion method of the Global Navigation Satellite System(GNSS)and lidar was explored.A low-cost,high-precision and easy-to-implement outdoor unmanned vehicle positioning and mapping scheme is studied.The main research work is as follows.First of all,in view of the difficulty and complexity of multi-sensor data processing,the time and space synchronization of multi-sensor data is first realized.According to the measurement principle of sensors such as lidar and integrated navigation,the internal and external parameters are calibrated,and the time synchronization of the data is realized according to the time stamp.Then the original point cloud data is processed to obtain effective data.Secondly,aiming at the problem that the traditional laser odometer cannot eliminate the accumulated error in large-scale scenes,and the built-up point cloud map ghost phenomenon is severely unable to be used for positioning,a multi-constrained laser SLAM solution is proposed.The front end adopts a feature matching odometer based on principal component analysis,and the back end adds prior environmental information constraints such as GNSS and ground plane to optimize the cumulative error of the laser odometer.The feature closed-loop constraint corrects the historical observation information,eliminates the ghost phenomenon,and improves the map accuracy.Experiments show that has better effect in long-distance trajectory positioning and mapping.Finally,aiming at the environmental cognition of unmanned vehicles in complex scenarios,a multi-layer map environment expression form for unmanned vehicles is proposed,and an auxiliary positioning method is designed to solve the problem that outdoor positioning of unmanned vehicles relies heavily on GNSS signals.Construct an outdoor multi-level environment map based on the optimal trajectory obtained by SLAM,and then use GNSS to provide the initial coordinate position,and then segment the point cloud map according to the coordinates to realize the dynamic loading of the point cloud map in blocks,and the current frame obtained by the lidar in real time matching with the local map will finally realize the positioning and navigation of unmanned vehicles.The experimental results show that this method has a good effect in the positioning of outdoor unmanned vehicles.